The DOI matches its Crossref record; 75 of the 75 references it cites were found in a registry or an index; 34 of the 36 sources its claims cite were checked for support.
An Accurate and Rapidly Calibrating Speech Neuroprosthesis
Open accessPublished version of a preprint
Abstract
BACKGROUND: Brain-computer interfaces can enable communication for people with paralysis by transforming cortical activity associated with attempted speech into text on a computer screen. Communication with brain-computer interfaces has been restricted by extensive training requirements and limited accuracy. METHODS: A 45-year-old man with amyotrophic lateral sclerosis (ALS) with tetraparesis and severe dysarthria underwent surgical implantation of four microelectrode arrays into his left ventral precentral gyrus 5 years after the onset of the illness; these arrays recorded neural activity from 256 intracortical electrodes. We report the results of decoding his cortical neural activity as he attempted to speak in both prompted and unstructured conversational contexts. Decoded words were displayed on a screen and then vocalized with the use of text-to-speech software designed to sound like his pre-ALS voice. RESULTS: On the first day of use (25 days after surgery), the neuroprosthesis achieved 99.6% accuracy with a 50-word vocabulary. Calibration of the neuroprosthesis required 30 minutes of cortical recordings while the participant attempted to speak, followed by subsequent processing. On the second day, after 1.4 additional hours of system training, the neuroprosthesis achieved 90.2% accuracy using a 125,000-word vocabulary. With further training data, the neuroprosthesis sustained 97.5% accuracy over a period of 8.4 months after surgical implantation, and the participant used it to communicate in self-paced conversations at a rate of approximately 32 words per minute for more than 248 cumulative hours. CONCLUSIONS: In a person with ALS and severe dysarthria, an intracortical speech neuroprosthesis reached a level of performance suitable to restore conversational communication after brief training. (Funded by the Office of the Assistant Secretary of Defense for Health Affairs and others; BrainGate2 ClinicalTrials.gov number, NCT00912041.).
Overview
CoChat Summary
TL;DR
An intracortical speech neuroprosthesis in a 45-year-old man with ALS and severe dysarthria decoded attempted speech into text with high accuracy after brief training. It achieved 99.6% accuracy with a 50-word vocabulary on the first day of use and 90.2% accuracy with a 125,000-word vocabulary on the second day, then sustained 97.5% accuracy during self-paced conversations over 8.4 months.
Key findings
- Twenty-five days after implantation, after 30 minutes of cortical recording collection, the neuroprosthesis achieved 99.6% accuracy with a 50-word vocabulary; on the second day, after 1.4 additional hours of training, it achieved 90.2% accuracy with a 125,000-word vocabulary.
- With further training, the neuroprosthesis sustained 97.5% accuracy for self-paced conversations over 248 cumulative hours across 8.4 months after implantation.
- In selected Conversation Mode sessions, word error rate was 3.7% (95% CI, 3.3% to 4.3%) across 925 sentences with known true labels; the participant self-reported that 52.9% and 32.3% of all Conversation Mode sentences were decoded correctly or mostly correctly, respectively.
- Average Copy Task decoding in the final 5 evaluation sessions had a word error rate of 2.5% (95% CI, 2.0% to 3.1%) at a self-paced speaking rate of 31.6 words per minute (95% CI, 31.2% to 32.0%).
Methods
- Single-participant study: a 45-year-old left-handed man with ALS, tetraparesis, and severe dysarthria underwent implantation of four microelectrode arrays into the left precentral gyrus, recording from 256 intracortical electrodes.
- Data were collected in 84 sessions over 32 weeks in the participant’s home, using an instructed-delay Copy Task and a self-paced Conversation Mode; decoding occurred in real time, with phoneme predictions every 80 ms and language models assembling phonemes into words and sentences.
- Performance was evaluated with phoneme error rate and word error rate; confidence intervals were estimated by randomly resampling each dataset 10,000 times with replacement and were not adjusted for multiplicity.
Limitations
- The paper states that the study involved a single participant and it is uncertain if similar results can be expected in future users.
- The paper states that the durability of the system as ALS progresses has not been extensively studied and that the authors cannot comment on use of this system in other disorders.
- The text makes evident that confidence intervals were not adjusted for multiplicity, and the first-ever closed-loop block was excluded from evaluation because the participant cried with joy when words appeared correctly on-screen.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
13 Valid · 3 Not valid · 2 Not checked · 43 Own results and interpretations
Some cited sources support this claim and some don’t — read each source’s verdict below.
In the paper
“Efforts to develop a speech neuroprosthesis are built largely on studies using data that are retrospectively analyzed from able-bodied speakers undergoing electrophysiological monitoring for clinical purposes 7-16 .”
What the sources say
- Supportedcites [7] 7. Herff C, Heger D, de Pesters A, et al. Brain-to-text: decoding spoken phrases from phone representations in the brain. Front Neurosci [Internet] 2015 [cited 2023 Dec 11];8. Available from: https://www.frontiersin.org/articles/10.3389/fnins.2015.00217 source
This paper is an example of the kind of study cited. Materials and methods (§§2.1–2.3) describes speech recordings from seven epilepsy patients whose ECoG grids were implanted for presurgical clinical purposes; the Conclusion (§4.4) says the decoding evaluations were performed offline. The participants were speech-capable patients, rather than healthy volunteers. (LLM, conf=0.90)
- Partly supportedcites [8] 8. Kellis S, Miller K, Thomson K, Brown R, House P, Greger B. Decoding spoken words using local field potentials recorded from the cortical surface. J Neural Eng 2010;7(5):056007. [PubMed: 20811093] sourceExperimental — accuracy not yet measured
Methods §2.1 describes a patient undergoing ECoG monitoring for epilepsy, and §2.2 describes recorded spoken-word trials selected for subsequent decoding analysis. This paper therefore exemplifies speech-neuroprosthesis research using data from a speaking clinical patient. It does not establish that such studies constitute the majority of efforts, and its participant had medically refractory epilepsy rather than being unqualifiedly able-bodied. (LLM, conf=0.86)
- Supportedcites [9] 9. Mugler EM, Patton JL, Flint RD, et al. Direct classification of all American English phonemes using signals from functional speech motor cortex. J Neural Eng 2014;11(3):035015. [PubMed: 24836588] source
This paper is an example of the studies cited: Methods §2.1 describes four participants undergoing ECoG monitoring for treatment of seizures, and §2.2 says their speech recordings were synchronized with ECoG recorded on a clinical system. Methods §2.5 describes subsequent phoneme classification using cross-validation, while the Discussion distinguishes this analysis from a real-time speech BCI. The paper supports its use as an example, rather than independently establishing how large a share of the field uses this approach. (LLM, conf=0.86)
- Supportedcites [10] 10. Ramsey NF, Salari E, Aarnoutse EJ, Vansteensel MJ, Bleichner MG, Freudenburg ZV. Decoding spoken phonemes from sensorimotor cortex with high-density ECoG grids. NeuroImage 2018;180(Pt A):301-11. [PubMed: 28993231] source
The Introduction describes five patients with epilepsy who had ECoG grids implanted during evaluation for seizure surgery. The Methods (Overt phoneme production Task; Spatial-Temporal Cortical Activation Pattern Classification; pages not available in our copy) describe their spoken-phoneme recordings and subsequent leave-one-out analysis of recorded trials. The paper therefore supports its role as an example of the study type cited, rather than independently establishing how much of the field uses that approach. (LLM, conf=0.89)
- Supportedcites [11] 11. Anumanchipalli GK, Chartier J, Chang EF. Speech synthesis from neural decoding of spoken sentences. Nature 2019;568(7753):493-8. [PubMed: 31019317] source
The Abstract describes a speech-prosthesis study using recordings from five participants undergoing intracranial monitoring for epilepsy treatment while they spoke aloud. Methods > Participants and experimental task describes their spoken-sentence recordings, and Methods > Neural decoder describes training and testing on recorded sessions. This paper supports the cited sentence as an example of that approach; it does not itself establish how common the approach is across the field. Page numbers are unavailable in our copy. (LLM, conf=0.89)
- Partly supportedcites [12] 12. Moses DA, Leonard MK, Makin JG, Chang EF. Real-time decoding of question-and-answer speech dialogue using human cortical activity. Nat Commun 2019;10(1):3096. [PubMed: 31363096] sourceExperimental — accuracy not yet measured
Methods > Participants (page unavailable) confirms that the study recorded from patients who could speak and had ECoG arrays implanted for clinical seizure monitoring. But Methods > Modeling says the trained models were used during online testing, and the Discussion describes real-time decoding. This paper therefore supports the clinical-monitoring aspect, not the characterization of its data as retrospectively analyzed; it also does not establish that most efforts relied on such studies. (LLM, conf=0.84)
- Partly supportedcites [13] 13. Stavisky SD, Willett FR, Wilson GH, et al. Neural ensemble dynamics in dorsal motor cortex during speech in people with paralysis. eLife 2019;8:e46015. [PubMed: 31820736] sourceExperimental — accuracy not yet measured
The Introduction (page not provided) says earlier direct speech recordings were made during clinical treatment for epilepsy or Parkinson’s disease, supporting the clinical-recording part of the background claim. It does not establish that speech-neuroprosthesis efforts were built largely on retrospectively analyzed data from able-bodied speakers. The Introduction also identifies this study’s participants as people with tetraplegia, rather than able-bodied speakers. (LLM, conf=0.86)
- Could not verifycites [14] 14. Stavisky SD, Willett FR, Avansino DT, Hochberg LR, Shenoy KV, Henderson JM. Speechrelated dorsal motor cortex activity does not interfere with iBCI cursor control. J Neural Eng 2020;17(1):016049. [PubMed: 32023225] sourceExperimental — accuracy not yet measured
The Introduction discusses efforts to build speech BCIs, but does not say those efforts rely largely on retrospective analyses of recordings from able-bodied speakers undergoing clinical monitoring. The Methods describe a prospective research session with a participant with tetraplegia; searches for retrospective analyses and clinical monitoring did not establish the broader claim. (LLM, conf=0.73)
- Supportedcites [15] 15. Berezutskaya J, Freudenburg ZV, Vansteensel MJ, Aarnoutse EJ, Ramsey NF, Van Gerven MAJ. Direct speech reconstruction from sensorimotor brain activity with optimized deep learning models. J Neural Eng 2023;20(5):056010. source
The Introduction says speech-decoding research is primarily conducted with non-disabled participants who produce speech while their brain activity is recorded. It also describes preparatory research with epilepsy patients temporarily implanted with ECoG electrodes for clinical monitoring. The Methods (ECoG experiment) describe analyzing the recorded data afterward; no page numbers are available in our copy. (LLM, conf=0.91)
- Disputedcites [16] 16. Guenther FH, Brumberg JS, Wright EJ, et al. A Wireless Brain-Machine Interface for Real-Time Speech Synthesis. PLOS ONE 2009;4(12):e8218. [PubMed: 20011034] sourceExperimental — accuracy not yet measured
The Introduction describes a volunteer with locked-in syndrome who received an implanted electrode for this study, not an able-bodied speaker undergoing clinical monitoring. The Abstract’s Methodology/Principal Findings says his neural signals drove a speech synthesizer with immediate feedback, contrary to the claim of retrospective analysis. (LLM, conf=0.90)
Checked by a reasoning model — may be wrong.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
Some cited sources support this claim and some don’t — read each source’s verdict below.
In the paper
“and required approximately 17 hours of recording to collect sufficient training data to obtain that level of performance.”
What the sources say
- Disputedcites [19] 19. Metzger SL, Littlejohn KT, Silva AB, et al. A high-performance neuroprosthesis for speech decoding and avatar control. Nature 2023;620(7976):1037-46. [PubMed: 37612505] sourceExperimental — accuracy not yet measured
The Text decoding section reports about 1.6 hours of training data per day over 13 days—approximately 21 hours for this study—and describes that as a shorter collection period than its previous work. It does not support the claim that prior studies required approximately 17 hours; its comparison points in the opposite direction. Page numbers were unavailable in our copy. (LLM, conf=0.81)
Checked by a reasoning model — may be wrong.
- Partly supportedcites [21] 21. Willett FR, Kunz EM, Fan C, et al. A high-performance speech neuroprosthesis. Nature 2023;620(7976):1031-6. [PubMed: 37612500] sourceExperimental — accuracy not yet measured
In “Decoding attempted speech” (page unavailable), the paper reports collecting training recordings and achieving high decoding accuracy with a dataset of 10,850 sentences. It does not give a total of approximately 17 hours or establish that this amount of recording was required; “Design considerations for speech BCIs” reports good performance even without new-day training data. (LLM, conf=0.84)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Disputed
Every source the paper cites for this claim was checked and supports it.
In the paper
“Brain-computer interfaces can enable communication for people with paralysis by transforming cortical activity associated with attempted speech into text on a computer screen.”
- Also stated in: § Introduction: · p. 3 · plain
“Brain-computer interfaces are a promising communication technology that can directly decode the user's intended speech from cortical neural signals 6 .”
What the sources say
- Supportedcites [6] 6. Luo S, Rabbani Q, Crone NE. Brain-Computer Interface: Applications to Speech Decoding and Synthesis to Augment Communication. Neurotherapeutics 2022;19(1):263-73. [PubMed: 35099768] source
The Sentence Decoding section (page unavailable) reports online decoding from cortical recordings collected during attempted speech by a participant with quadriplegia and anarthria. It says the work demonstrated the feasibility of transcribing neural data into textual output, supporting the claim's description of communication through text. (LLM, conf=0.91)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Also stated in: § Introduction: · p. 3 · plain
Every source the paper cites for this claim was checked and supports it.
In the paper
“and 23.8% with a 125,000-word vocabulary 21”
What the sources say
- Supportedcites [21] 21. Willett FR, Kunz EM, Fan C, et al. A high-performance speech neuroprosthesis. Nature 2023;620(7976):1031-6. [PubMed: 37612500] source
The ‘Decoding attempted speech’ section reports a 23.8% word error rate with a 125,000-word vocabulary across vocalizing evaluation days. Table 1 identifies this as the online result for vocal speech. (LLM, conf=0.95)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“Our choice of targeting speech motor cortex was informed by our previous study in another aforementioned individual that found two arrays in 6v provided informative signals for speech decoding 21 .”
What the sources say
- Supportedcites [21] 21. Willett FR, Kunz EM, Fan C, et al. A high-performance speech neuroprosthesis. Nature 2023;620(7976):1031-6. [PubMed: 37612500] source
The Main section identifies T12 as the study participant and states that two arrays were placed in area 6v. The Decoding attempted speech section reports that both arrays were useful for decoding, with lower word error rates when their signals were combined. (LLM, conf=0.94)
plainstates an associationSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
33 more claims · 4 with their cited sources checked
Create a free account to see all 61 claims and every source verdict.
- Probably not a substantial claim. 16
- It may state two separate claims as one. 12
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“(although high specificity has also been demonstrated without such design2).”
From Inner speech in motor cortex and implications for speech neuroprostheses (2025) · cites this paper as [2]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“Speech brain-computer interfaces (BCIs) show promise in restoring communication to people with paralysis1–3”
From Inner speech in motor cortex and implications for speech neuroprostheses (2025) · cites this paper as [2]
The cited source says what the claim attributes to it.
In the citing paper
“This aligns with previous speech BCI studies in which the middle precentral gyrus3 and most ventral precentral gyrus1,2 demonstrated the greatest signal contributions to speech decoding.”
From Inner speech in motor cortex and implications for speech neuroprostheses (2025) · cites this paper as [2]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated major: it would touch a result — a value a result rests on, reported and recomputed p-values on opposite sides of the significance level, two passages that conflict with no sign which is right, or a cited work retracted before this paper was published.
In the paper
§ Abstract · p. 3
“With further training data, the neuroprosthesis sustained 97.5% accuracy for self-paced conversations for over 248 cumulative hours over 8.4 months after surgical implantation.”
§ Conversational speech using the brain-computer interface · p. 8
“The word error rate during selected Conversation Mode sessions was 3.7% (95% CI, 3.3% to 4.3%; Fig. 4b)”
Citations
Cites
NWChem: A comprehensive and scalable open-source solution for large scale molecular simulations
Valiev, M., Bylaska, E. J., Govind, N., Kowalski, K., Straatsma, T. P., Van Dam, H. J. J., Wang, D., Nieplocha, J., Apra, E., Windus, T. L., et al. (2010) · Computer Physics Communications · doi:10.1016/j.cpc.2010.04.018 · Cited by 5,779
A multi-modal parcellation of human cerebral cortex
Glasser, M. F., Coalson, T. S., Robinson, E. C., Hacker, C. D., Harwell, J., Yacoub, E., Ugurbil, K., Andersson, J., Beckmann, C. F., Jenkinson, M., et al. (2016) · Nature · doi:10.1038/nature18933 · Cited by 5,111
Self-citationUnbiased average age-appropriate atlases for pediatric studies
Fonov, V., Evans, A. C., Botteron, K., Almli, C. R., McKinstry, R. C., & Collins, D. L. (2011) · NeuroImage · doi:10.1016/j.neuroimage.2010.07.033 · Cited by 2,290
SWITCHBOARD: telephone speech corpus for research and development
Godfrey, J. J., Holliman, E. C., & McDaniel, J. (1992) · [Proceedings] ICASSP-92: 1992 IEEE International Conference on Acoustics, Speech, and Signal Processing · doi:10.1109/icassp.1992.225858 · Cited by 2,134
Unbiased nonlinear average age-appropriate brain templates from birth to adulthood
Fonov, V., Evans, A., McKinstry, R., Almli, C., & Collins, D. (2009) · NeuroImage · doi:10.1016/s1053-8119(09)70884-5 · Cited by 2,102
Salimi-Khorshidi, G., Douaud, G., Beckmann, C. F., Glasser, M. F., Griffanti, L., & Smith, S. M. (2014) · NeuroImage · doi:10.1016/j.neuroimage.2013.11.046 · Cited by 1,770
Self-citationHigh-performance brain-to-text communication via handwriting
Willett, F. R., Avansino, D. T., Hochberg, L. R., Henderson, J. M., & Shenoy, K. V. (2021) · Nature · doi:10.1038/s41586-021-03506-2 · Cited by 1,024
Self-citationSpeech synthesis from neural decoding of spoken sentences
Anumanchipalli, G. K., Chartier, J., & Chang, E. F. (2019) · Nature · doi:10.1038/s41586-019-1119-1 · Cited by 883
A brain-computer interface that evokes tactile sensations improves robotic arm control
Flesher, S. N., Downey, J. E., Weiss, J. M., Hughes, C. L., Herrera, A. J., Tyler-Kabara, E. C., Boninger, M. L., Collinger, J. L., & Gaunt, R. A. (2021) · Science · doi:10.1126/science.abd0380 · Cited by 617
Harms, M. P., Somerville, L. H., Ances, B. M., Andersson, J., Barch, D. M., Bastiani, M., Bookheimer, S. Y., Brown, T. B., Buckner, R. L., Burgess, G. C., et al. (2018) · NeuroImage · doi:10.1016/j.neuroimage.2018.09.060 · Cited by 584
Self-citation
Cited by
Silva, A. B., Littlejohn, K. T., Liu, J. R., Moses, D. A., & Chang, E. F. (2024) · Nature Reviews Neuroscience · doi:10.1038/s41583-024-00819-9 · Cited by 103
An instantaneous voice-synthesis neuroprosthesis
Wairagkar, M., Card, N. S., Singer-Clark, T., Hou, X., Iacobacci, C., Miller, L. M., Hochberg, L. R., Brandman, D. M., & Stavisky, S. D. (2025) · Nature · doi:10.1038/s41586-025-09127-3 · Cited by 146
Self-citationA streaming brain-to-voice neuroprosthesis to restore naturalistic communication
Littlejohn, K. T., Cho, C. J., Liu, J. R., Silva, A. B., Yu, B., Anderson, V. R., Kurtz-Miott, C. M., Brosler, S., Kashyap, A., Hallinan, I. P., et al. (2025) · Nature Neuroscience · doi:10.1038/s41593-025-01905-6 · Cited by 130
Willsey, M. S., Shah, N. P., Avansino, D. T., Hahn, N. V., Jamiolkowski, R. M., Kamdar, F. B., Hochberg, L. R., Willett, F. R., & Henderson, J. M. (2025) · Nature Medicine · doi:10.1038/s41591-024-03341-8 · Cited by 116
Self-citationDecoding the brain: From neural representations to mechanistic models
Mathis, M. W., Rotondo, A. P., Chang, E. F., Tolias, A. S., & Mathis, A. (2024) · Cell · doi:10.1016/j.cell.2024.08.051 · Cited by 128
Metadata
- Authors
- Card, N. S., Wairagkar, M., Iacobacci, C., Hou, X., Singer-Clark, T., Willett, F. R., Kunz, E. M., Fan, C., Vahdati Nia, M., Deo, D. R., Srinivasan, A., Choi, E. Y., Glasser, M. F., Hochberg, L. R., Henderson, J. M., Shahlaie, K., Stavisky, S. D., & Brandman, D. M.
- Affiliations
- Howard Hughes Medical Institute (US)
- From the Departments of Neurological Surgery (N.S.C., M.W., C.I., X.H., T.S.-C., M.V.N., A.S., K.S., S.D.S., D.M.B.), Computer Science (X.H., M.V.N.), and Biomedical Engineering (T.S.-C., A.S.), University of California, Davis, Davis, and the Departments of Neurosurgery (D.R.D., E.Y.C., J.M.H.), Electrical Engineering (E.M.K.), and Computer Science (C.F.), the Wu Tsai Neurosciences Institute (E.M.K., J.M.H.), the Howard Hughes Medical Institute (F.R.W.), and Bio-X (J.M.H.), Stanford University, Stanford — both in California; the Departments of Radiology and Neuroscience, Washington University School of Medicine, Saint Louis (M.F.G.); the School of Engineering and Carney Institute for Brain Sciences, Brown University (L.R.H.), and the Center for Neurorestoration and Neurotechnology, Department of Veterans Affairs Office of Rehabilitation Research and Development, VA Providence Healthcare (L.R.H.) — both in Providence, RI; and the Center for Neurotechnology and Neurorecovery, Department of Neurology, Massach...
- ORCID
- Venue
- New England Journal of Medicine
- Publisher
- Massachusetts Medical Society
- Published
- 2024-08-15
- Volume
- 391
- Issue
- 7
- Pages
- 609-618
- ISSN
- 0028-4793 (ISSN-L)
- 1533-4406
- PMC
- PMC11328962
- OpenAlex
- W4401560881
- Semantic Scholar
- cf30ed6b05da4ce9cd70550a9532a2cb3ad6db49
- Cited by
- 268
- References
- 32
- Citations by year
- 2026: 128 · 2025: 115 · 2024: 24
- FWCI
- 50.11
- Type
- journal article
- Language
- en
- Open access
- green
- doi.org
- Licence
- Locations
- New England Journal of Medicine (publishedVersion)
- medRxiv (submittedVersion, cc-by-nc-nd)
- PubMed (publishedVersion)
- PubMed Central (acceptedVersion)
- eScholarship (California Digital Library) (submittedVersion)
- PubMed Central (submittedVersion, cc-by)
- Digital Commons@Becker (Washington University School of Medicine) (submittedVersion)
- Funders
- National Institute on Deafness and Other Communication Disorders (DP2DC021055, U01DC019430, U01DC17844)
- A.P. Giannini Foundation Postdoctoral Fellowship
- Simons Foundation (872146SPI)
- Howard Hughes Medical Institute
- Burroughs Wellcome Fund
- Office of the Assistant Secretary of Defense for Health Affairs (AL220043)
- University of California, Davis
- Searle Scholars Program
- U.S. Department of Veterans Affairs (A2295-R)
- Stanford Wu Tsai Neurosciences Institute
- Topics
- Life Sciences › Neuroscience › Cognitive Neuroscience › EEG and Brain-Computer Interfaces
- Life Sciences › Neuroscience › Cellular and Molecular Neuroscience › Neuroscience and Neural Engineering
- Physical Sciences › Computer Science › Human-Computer Interaction › Gaze Tracking and Assistive Technology
- Keywords
- brain-computer interface, amyotrophic lateral sclerosis, tetraparesis, intracortical microelectrode arrays, text-to-speech, conversational communication
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: An Accurate and Rapidly Calibrating Speech Neuroprosthesis. https://app.cochat.ai/papers/10.1056/nejmoa2314132
- The paper
- Card, N. S., Wairagkar, M., Iacobacci, C., Hou, X., Singer-Clark, T., Willett, F. R., Kunz, E. M., Fan, C., Vahdati Nia, M., Deo, D. R., Srinivasan, A., Choi, E. Y., Glasser, M. F., Hochberg, L. R., Henderson, J. M., Shahlaie, K., Stavisky, S. D., & Brandman, D. M. (2024). An Accurate and Rapidly Calibrating Speech Neuroprosthesis. New England Journal of Medicine. https://doi.org/10.1056/nejmoa2314132
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.